Benjamin Solnik
Research Areas
Authored Publications
Sort By
Bayesian Optimization for a Better Dessert
Subhodeep Moitra
Proceedings of the 2017 NIPS Workshop on Bayesian Optimization, December 9, 2017, Long Beach, USA (to appear)
Preview abstract
We present a case study on applying Bayesian Optimization to a complex real-world system; our challenge was to optimize chocolate chip cookies. The process was a mixed-initiative system where both human chefs, human raters, and a machine optimizer participated in 144 experiments. This process resulted in highly rated cookies that deviated from expectations in some surprising ways -- much less sugar in California, and cayenne in Pittsburgh. Our experience highlights the importance of incorporating domain expertise and the value of transfer learning approaches.
View details
Google Vizier: A Service for Black-Box Optimization
Subhodeep Moitra
ACM (2017), pp. 1487-1495
Preview abstract
Any sufficiently complex system acts as a black box when it becomes easier to
experiment with than to understand. Hence, black-box optimization has become
increasingly important as systems have become more complex. In this paper we
describe Google Vizier, a Google-internal service for performing
black-box optimization that has become the de facto parameter tuning
engine at Google. Google Vizier is used to optimize many of our machine
learning models and other systems, and also provides core capabilities to
Google's Cloud Machine Learning HyperTune subsystem. We discuss our
requirements, infrastructure design, underlying algorithms, and advanced
features such as transfer learning and automated early stopping that the
service provides.
View details
